- Email: [email protected]
New helio-photocatalytic–photovoltaic hybrid system for simultaneous water decontamination and solar energy conversion
Solar Energy 79 (2005) 353–359 www.elsevier.com/locate/solener
New helio-photocatalytic–photovoltaic hybrid system for simultaneous water decontamination and solar energy conversion Victor Sarria c, Simeon Kenfack a, Sixto Malato b, Julian Blanco b, Cesar Pulgarin a,* a
Ecole Polytechnique Fe´de´rale de Lausanne (EPFL), Laboratoire de Biotechnologie Environementale, CH-1015 Lausanne, Switzerland b Plataforma Solar de Almeria, Tabernas, 04200 Almeria, Spain c Departamento de Quı´mica, Universidad de los Andes, Cra. 1 No.18a-10, Bogota´, Colombia Received 21 July 2004; received in revised form 8 December 2004; accepted 23 February 2005 Available online 17 May 2005 Communicated by: Associate Editor Jaime Gimenez
Abstract Test results for a designed and installed hybrid photocatalytic–photovoltaic system (HPPS) are presented in this paper. The HPPS consists of one device with dual functions: The photocatalytic system uses UV radiation to promote degradation of organic pollutants, and beside it absorbs the IR radiation. The photovoltaic (PV) system converts the visible radiation into electricity, which can either be used directly by the recirculation pump or stored in a battery for other purposes. The suggested design aims to achieve an autonomous, environmentally friendly method for the treatment of biorecalcitrant pollutants. Two prototypes were erected: HPPSP, using Plexiglas and HPPSG using commercial glass. Both were tested outdoors to determine their impact on photovoltaic power production. Test results showed that PV power diminished to 14% and 22% for HPPSG and HPPSP respectively, compared to a PV panel alone. 2005 Elsevier Ltd. All rights reserved. Keywords: Photodegradation; Solar energy; Water treatment
1. Introduction The most common solar energy applications are thermal, photovoltaic (PV), and photochemical energy conversion. Over the last few years, there has been a growing interest in photovoltaic/thermal collectors
* Corresponding author. Tel.: +41 21 693 4720; fax: +41 21 693 4722. E-mail address: cesar.pulgarin@epfl.ch (C. Pulgarin).
(Sorensen and Munro, 2000), which can provide both electrical and thermal energy from the same system. Nevertheless, there are no reports of photovoltaic/photocatalytic hybrid systems. Fig. 1 (King and Kratochvil, 1997) shows the spectral response of a typical crystalline silicon (c-Si) photovoltaic module compared to the solar spectral irradiance. About 40% of the solar energy incident on a c-Si cell passes through it, because its energy is lower than that of the c-Si band gap. However a major part is absorbed by the sealing and other protective materials used in the
0038-092X/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.solener.2005.02.022
354
V. Sarria et al. / Solar Energy 79 (2005) 353–359
Nomenclature HPPS hybrid photocatalytic–photovoltaic system HPPSG hybrid photocatalytic–photovoltaic system using commercial glass HPPSP hybrid photocatalytic–photovoltaic system using plexiglas
1.8 1.6
Irradiance, (W m-2 nm-1)
1.4 Solar spectra
1.2 1 0.8 0.6 PV cell
0.4 0.2 0 305
410
570
780
980
1395
1800
2537
Wavelength (nm)
Fig. 1. Relative spectral response of a typical c-Si module compared to total solar spectral irradiance (King and Kratochvil, 1997).
module. This may increase the temperature of the module to around 80 C under hot humid conditions. Warming of the PV module can adversely affect its performance due to deterioration of its opto-electronic properties. Moreover, it is known (Custodio et al., 2001) that the short circuit current (Isc) increases with the temperature, whereas the open circuit voltage (Voc) decreases in the same conditions. Generally, the decrease in voltage is much more pronounced than the increment in current (Radziemska, 2003). From a microscopic point of view, the temperature increment in the PV cell causes decreased lifetime and diffusion length of the minority carriers, as well as a decrease in mobility of the majority carriers. Also, when carriers are generated on both sides of the junction, both Fermi levels tend toward the intrinsic level, thus decreasing the potential barrier. The decrease in barrier potential causes an increase in short circuit current and a decrease in open circuit voltage. Efficiency usually falls 30% when the temperature of the cell increases from 20 to 60 C.
PV UV VIS IR CPC
photovoltaic system ultraviolet solar radiation visible solar radiation infrared solar radiation compound parabolic collector
Custodio et al. (2001) had designed an infrared water filter to absorb the part of the solar spectrum, which cannot be exploited by the photovoltaic module and is detrimental for the ideal performance of the module. This filter consisted of a layer of purified water, which is almost transparent to that part of the solar spectrum (400–800 nm) to which c-Si is highly sensitive, and highly absorbent for infrared wavelengths outside of this spectral range. Photocatalytic systems: In recent years, solar UV/visible radiation has been successfully applied for water treatment in the context of so called Advanced Oxidation Processes (AOPs) (Ollis and Ekabi, 1993). These methods use the interaction between ultraviolet radiation and a catalyst or oxidant, to generate OH radicals in the bulk of the solution. OH radicals have a very high oxidation potential and are able to degrade organic compounds present in water, generating non-toxic, biodegradable intermediates that can be removed in a subsequent biological treatment process (Sarria et al., 2002, 2003). Blake (2001) has compiled a survey of literature describing photocatalysis for removal of hazardous compounds from water and air listing 941 scientific papers that describe a variety of process applications. Among the AOPs, the photo-assisted Fenton Fe(III)/ light system, without addition of any electron acceptor other than O2 from air, has received special attention in earlier work, as a potential wastewater pre-treatment (Mailhot et al., 2002). Previous research has shown that under UV irradiation, iron(III) salts can promote the photooxidation of organic compounds (Mailhot et al., 2000; Mazellier and Bolte, 2001; Rodriguez et al., 2002). The agent responsible for these reactions could be the hydroxyl radical formed by photochemical dissociation of Fe(III)-hydroxy complex (Fe(OH)2+). The reaction can be simply expressed as (Faust and Hoigne, 1990): FeðOHÞ2þ þ hm ! Fe2þ þ OH. Since the silicon PV panel converts mainly visible solar radiation into electricity, and since the photocatalytic degradation needs near UV radiation, the approach of simultaneously employing both types of radiation in one engine and with no additional surface can be advantageous.
V. Sarria et al. / Solar Energy 79 (2005) 353–359
Consequently, the aim of this work was to develop a hybrid photocatalytic–photovoltaic system (HPPS). The photocatalytic system uses the UV radiation to promote the degradation of organic pollutants and absorbs the IR radiation, thereby protecting the PV panels and lengthening their lifetime. The photovoltaic system converts sunlight into electricity which can be transformed to alternating current for powering the recirculation pump. It could also be accumulated by a battery for other purposes. This new concept can therefore represent an autonomous and environmentally friendly method for the treatment of biorecalcitrant pollutants. In this paper, we report the construction and the results of basic batch experiments using the HPPS. Photocatalytic degradation of a model biorecalcitrant compound, 5-amino-6-methyl-2-benzimidazole (AMBI) was performed, and the efficiency of the HPPS treatment was compared with a compound parabolic solar collector. In order to validate our results, each experiment was performed three times, and standard deviation was calculated and reported for Table 2 and Figs. 5 and 6.
2. Experimental section 2.1. Hybrid photocatalytic–photovoltaic system (HPPS) A schematic representation of the new HPPS is presented in Fig. 2. Solar radiation comprising the UV, Vis and IR radiations, reaches the photochemical reactor which uses the UV radiation for degradation of the organic pollutants. The IR radiation is filtered by the water in the photocatalytic reactor and only the visible
355
radiation, useful for the production of electrical energy, reaches the PV panel situated below the photocatalytic reactor. Fig. 3 shows a picture of the HPPS installed at the EPFL—Switzerland. It is made up of 4 modules (collector surface, 1.5 m2, photoreactor volume 7 l, and total reactor volume 25 l) mounted on a fixed platform inclined 46 (local latitude). Each module consists of one photocatalytic reactor and one PV panel. The photocatalytic system consists of 4 rectangular cells made of a UV/Vis transparent material through which the water to be treated recirculates. These cells, placed over the PV panels, are connected in series and the water flows directly from one to another and finally to a tank equipped with a centrifugal recirculation pump. The photovoltaic system, built in collaboration with Solstist SA., is composed of the following parts: • 4 PV panels, model BP 270 F (50-W) connected in series; • A sinewave inverter (Studer Joker 400), which converts the low voltage DC to 220-V AC, and stops energy accumulation in the battery; • A battery (12 V 105 Ah) to store the energy; • A Voltmeter to measure the voltage in the battery; • A battery charger, as the solar energy is insufficient for a practical application; • Two electricity meters, to enable the difference between the electricity from the PV panels to be compensated from the local electricity grid; • Two chronometers, for timing the PV/local grid power input;
Fig. 2. Schematic representation of the HPPS.
356
V. Sarria et al. / Solar Energy 79 (2005) 353–359
Fig. 3. Photo of the HPPS installed at the EPFL—Switzerland.
• A three-position commuter (1. automatic, 0. local electricity network, and 2. solar PV panels). 2.2. Compound parabolic collector The low concentrating CPC collector (with a concentration ratio of about 1), described in a previous publication (Sarria et al., 2003) consists of three 4-tube modules (photoreactor material: Pyrex, collector surface: 1.5 m2, photoreactor volume: 12 l, and total reactor volume: 25 l). The three reactor modules are mounted on a fixed platform inclined 46 (local latitude) and connected in series with water flowing directly through them at 30 l min1, leading finally to a recirculation tank connected to a centrifugal pump. Solar ultraviolet radiation (UV), which is a highly significant parameter for the correct treatment of data obtained from solar photocatalytic experiments, was determined during the experiments. A global UV KIPP & ZONEN, model CUV3 radiometer mounted on a 46 fixed-angle platform (the same angle as the CPCs) was used to provide data on global solar energy incident per unit area, WUV m2 (UVG,n). Variation in solarUV power during experiments is made still more noticeable by clouds. With Eq. (1), combination of data from several days experiments and their comparison with other photocatalytic experiments is possible. QUV;n ¼ QUV;n1 þ Dtn UVG;n
Areactor V TOT
ð1Þ
where Dtn = tntn1, tn is the experimental time for each sample, UVG;n is the average UVG during Dtn, Areactor is the irradiated surface of collectors CPC or HPPS (1.5 m2), VTOT is the total volume of the photoreactor (25 l), and QUV,n is the accumulated incident energy on the reactor for each sample during the experiment per unit of volume (kJ l1).
2.3. Reagents 5-amino-6-mehtyl-2-benzimidazolone (AMBI), a previously described (Sarria et al., 2002) biorecalcitrant dye (C8H9N3O), was chosen to test photocatalytic degradation. FeCl3 Æ 6H2O was purchased from Fluka and was used without further purification. The chemicals purchased to assist in HPLC analyses were obtained from Fluka. Milli-Q water was used throughout for preparation of aqueous solutions or as a component of the water-acetonitrile (HPLC grade) mobile phase in HPLC analysis. 2.4. Chemical analyses High performance liquid chromatography (HPLC) was carried out in a Varian 9065 unit using a diode array detector between 200 and 400 nm. A Spherisorb silica column ODS-2 was used and the mobile phase (flow: 1 ml min1) consisted of an acetonitrile/water mixture (50%/50% v/v). This technique allows measuring the AMBI concentration in solution. 3. Results 3.1. Transmittance of different materials suitable for manufacture the HPPS To optimize the performance of the HPPS, the photocatalytic reactor must be made of highly UV-transmissive material. Fig. 4 shows the UV/Vis transmissivity spectra for different materials. Quartz and Pyrex have the highest UV transmissivity; but are not suitable for this purpose, due to their high cost. Commercial glass and Plexiglas, on the other hand, are economically suitable and also resistant to pressure and heating during outdoor operation.
V. Sarria et al. / Solar Energy 79 (2005) 353–359
radiation as shown on the right of Fig. 5. The power is observed to be only slight diminished by the photocatalytic reactor: 14% and 22% by the HPPSG and HPPSP, respectively. Another important parameter of the PV panel is the conversion efficiency. The average efficiency calculated for the PV panel is about 5.8% under normal operative conditions (without photoreactor) and about 4.89% and 4.67% for HPPSG and HPPSP. The above shows that the energy conversion of the PV panel is not significantly diminished by the installation of a photocatalytic reactor over the PV panel.
100 90
Quartz
Transmissivity (%)
80 70
Pyrex
60 50 Commercial glass
40 30 20
Plexiglas
10 0 200
300
400
500
600
357
700
800
3.3. Comparison of a CPC and the new HPPS system for the photocatalytic degradation of AMBI
Wavelength (nm) Fig. 4. Transmissivity of different materials suitable for manufacture of the HPPS.
Two photocatalytic reactors, one using commercial glass (HPPSG) and the other using Plexiglas XT (HPPSP) were fabricated for these experiments. 3.2. Photovoltaic power The sunlight-to-electric conversion parameter to be maximized is the power, which is the product of: current (I) · voltage (V) measured in a photovoltaic module. Fig. 5 shows the photovoltaic power of a PV panel, with and without glass and Plexiglas photocatalytic reactors as a function of the local time between 12:00 and 17:00. Power decreases as time increases, which is to be expected taking into account the diminishing solar
25
Power (W)
20
400
irradiance
HPPSG
300
HPPSP
15
200
10
Irradiance (W m-2)
PV
100
5 0 h0 12
0 h0 13
0 h0 14
0 h0 15
0 h0 16
0 h0 17
Local time
Fig. 5. PV panel power: (PV) panel alone, (HPPSG) with a commercial glass photoreactor, and (HPPSP) with a Plexiglas photoreactor in the new HPPS system. Comparison with the solar irradiance at the EPFL—Switzerland, October 7th 2002.
In order to evaluate the photocatalytic performance of the new HPPS, we simultaneously performed two degradation experiments using a well-known biorecalcitrant pollutant, AMBI (Sarria et al., 2001), one in the new HPPS reactor and another in the CPC, which until now has been demonstrated to be the best solar photocatalytic reactor for degradation applications (Malato et al., 2002), and compared the results. A diluted solution of AMBI was degraded by photoassisted Fenton using the CPC, the HPPSG and the HPPSP. The main characteristics of the reactors are presented in Table 1. Both the glass and Plexiglas HPPS reactors have the same dimensions. All three systems have the same illuminated surface. The main difference between these systems is the recirculation rate, which is higher in the CPC. The transversal section for photoreactors in the CPC and HPP systems are different. For that reason, the water flow for both systems was experimentally estimated using the ‘‘stop watch and bucket method’’. Basically, it was recorded the time to deliver a known volume, then volume divided by time is the flow rate. However, it was decided to go further in the exploratory evaluation of system performance, comparing the same illuminated surface. Fig. 6 shows ln(C/C0) evolution of the AMBI solution as a function of accumulated energy, calculated using Eq. (1), during the photocatalytic degradation experiments with the 3 solar reactors. C is the concentration during the experiment and C0 is the initial
Table 1 Characteristics of the compound parabolic collector (CPC) and the hybrid photovoltaic photocatalytic system (HPPS) Reactor Illuminated Illuminated Total Recirculation area (m2) volume (l) volume (l) (l min1) CPC HPPS
1.5 1.5
12 7
25 25
30 0.8
358
V. Sarria et al. / Solar Energy 79 (2005) 353–359 0.4 0.35 CPC
-ln (C/Co)
0.3 0.25
HPPSG
0.2 0.15
HPPSP
0.1
y = 0.001x
0.05 0 0
25
50
75
100
125
Quv(kJ l-1) Fig. 6. Comparison of the CPC and the HPPS systems.
Table 2 Comparison of the photodegradation constant (kapp) of AMBI solutions, using the HPPS and CPC reactors System
HPPSP
HPPSG
CPC
kapp (l kJ1)
0.0012 (±0.0004)
0.0021 (±0.0005)
0.0031 (±0.0003)
concentration of AMBI at the beginning of the experiment. According to the results presented in Fig. 6, it is clear that the CPC is the more performant from the kinetic point of view, however the HPP systems are energy autonomous. The constant of photodegradation (kapp) is obtained by calculating the slope of the curves in Fig. 6. Table 2 illustrates the kapp obtained for each reactor. Under the experimental conditions, the CPC is 1.5 and 3.1 times more efficient than the HPPSG and HPPSP, respectively for the treatment of a diluted solution of AMBI. However, more research will be necessary to find out whether the HPPS efficiency can be improved. The parameters involved should be, for instance, another kind of UV-transmissive material, agitation in the HPPS water circuit, water recirculation rate, and the use of an immobilized catalyst.
4. Conclusions A new hybrid photocatalytic photovoltaic system (HPPS) was developed, installed and operated at the EPFL in Switzerland. This HPPS uses solar UV radiation to evoke the photocatalytic degradation of organic pollutants, and visible radiation to generate electricity needed by the recirculation pump. HPPS represent an autonomous and environmental friendly approach for the treatment of biorecalcitrant pollutants. Commercial glass and Plexiglas, which have a good transmittance-cost ratio, were used to manufacture two
HPPS prototypes, HPPSP (Plexiglas) and HPPSG (commercial glass). The photovoltaic power of the PV panel was compared with and without the glass and Plexiglas photocatalytic reactors. The power output of the cell is almost directly proportional to the intensity of the sunlight. It was also observed that the power is only slightly diminished by addition of the photocatalytic reactor: 14% and 22% with the HPPSG and HPPSP, respectively. Degradation of the model compound, 5-amino-6methyl-2-benzimidazol (AMBI) is possible using photo-assisted Fenton. Under the experimented conditions, the CPC was shown to be 1.5 and 3.1 times more efficient than the HPPSG and HPPSP, respectively, for the degradation of AMBI. Acknowledgements This work was funded by OFES Contract no: 01.0433 under the European Community CADOX Research Project Contract no: EUK1-CTIC1-CT-200200122. We wish to thank Pascal Afolter of ‘‘Solstis SA’’. V. Sarria thanks ‘‘Comite´ de Investigaciones y Posgrados de la Facultad de Ciencias, Universidad de los Andes’’ for the financial support. We also wish to express our gratitude to the ‘‘Project environment EPFLUNIVALLE’’ and the skillful technical assistance of Jean-Pierre Kradolfer. The authors also wish to thank Mrs. Deborah Fuldauer for correcting the English. References Blake, D., 2001. Bibliography of work on the heterogeneous photocatalytic removal of hazardous compounds from water and air. Update number 4 to October 2001, NREL/ TP-510-31319. National Technical Information Service (NTIS), U.S. Department of Commerce, Springfield, VA 22161. Custodio, E., Acosta, L., Sebastian, P.J., Campos, J., 2001. A better solar module performance obtained by employing an infrared water filter. Solar Energy Materials and Solar Cells 70, 395–399. Faust, B.C., Hoigne, J., 1990. Photolysis of Fe(III)-hydroxy complexes as sources of OH radicals in clouds, fog and rain. Atmos. Environ. 24A, 79–89. King, D.L., Kratochvil, J.A., 1997. Measuring solar spectral and angle-of-incidence effects on photovoltaic modules and solar irradiance sensors. In: Proceedings of the 26th IEEE Photovoltaic Specialists Conference. Sandia National Laboratories, Anaheim, California. Mailhot, G., Asif, A., Bolte, M., 2000. Degradation of sodium 4-dodecylbenzenesulphonate photoinduced by Fe(III) in aqueous solution. Chemosphere 41, 363–370. Mailhot, G., Sarakha, M., Lavedrine, B., Caceres, J., Malato, S., 2002. Fe(III)-solar light induced degradation of diethyl phthalate (DEP) in aqueous solutions. Chemosphere 49, 525–532.
V. Sarria et al. / Solar Energy 79 (2005) 353–359 Malato, S., Blanco, J., Vidal, A., Fernandez, P., Caceres, J., Trincado, P., Oliveira, J.C., Vincent, M., 2002. New large solar photocatalytic plant: set-up and preliminary results. Chemosphere 47, 235–240. Mazellier, P., Bolte, M., 2001. 3-Chlorophenol elimination upon excitation of dilute iron(III) solution: evidence for the only involvement of Fe(OH)2+. Chemosphere 42, 361–366. Ollis, D.F., Ekabi, H.A. (Eds.), 1993. Photocatalytic Purification and Treatment of Water and Air. Elsevier, Amsterdam. Radziemska, E., 2003. The effect of temperature on the power drop in crystalline silicon solar cells. Renewable Energy 28, 1–12. Rodriguez, M., Abderrazik, N., Contreras, S., Chamarro, E., Gimenez, J., Esplugas, S., 2002. Iron(III) photoxidation of organic compounds in aqueous solutions. Appl. Catal. B. 37, 131–137.
359
Sarria, V., Parra, S., Invernizzi, M., Pe´ringer, P., Pulgarin, C., 2001. Photochemical–Biological Treatment of a Real Industrial Biorecalcitrant Wastewater Containing 5-amino-6methyl-2-benzimidazolone. Water Sci. Technol. 44, 93–101. Sarria, V., Parra, S., Adler, N., Pe´ringer, P., Pulgarin, C., 2002. Recent developments in the coupling of photoassisted and aerobic biological processes for the treatment of biorecalcitrant compounds. Catal. Today. 76, 301–315. Sarria, V., Deront, M., Peringer, P., Pulgarin, C., 2003. Degradation of a biorecalcitrant dye precursor present in industrial wastewaters by a new integrated iron(III) photoassisted-biological treatment. Appl. Catal. B. 40, 231– 246. Sorensen, H., Munro, D., 2000. Hybrid PV/T Collectors. In: Proceedings of the 2nd World Solar Electric Buildings Conference, Sydney, Australia.
New helio-photocatalytic–photovoltaic hybrid system for simultaneous water decontamination and solar energy conversion Victor Sarria c, Simeon Kenfack a, Sixto Malato b, Julian Blanco b, Cesar Pulgarin a,* a
Ecole Polytechnique Fe´de´rale de Lausanne (EPFL), Laboratoire de Biotechnologie Environementale, CH-1015 Lausanne, Switzerland b Plataforma Solar de Almeria, Tabernas, 04200 Almeria, Spain c Departamento de Quı´mica, Universidad de los Andes, Cra. 1 No.18a-10, Bogota´, Colombia Received 21 July 2004; received in revised form 8 December 2004; accepted 23 February 2005 Available online 17 May 2005 Communicated by: Associate Editor Jaime Gimenez
Abstract Test results for a designed and installed hybrid photocatalytic–photovoltaic system (HPPS) are presented in this paper. The HPPS consists of one device with dual functions: The photocatalytic system uses UV radiation to promote degradation of organic pollutants, and beside it absorbs the IR radiation. The photovoltaic (PV) system converts the visible radiation into electricity, which can either be used directly by the recirculation pump or stored in a battery for other purposes. The suggested design aims to achieve an autonomous, environmentally friendly method for the treatment of biorecalcitrant pollutants. Two prototypes were erected: HPPSP, using Plexiglas and HPPSG using commercial glass. Both were tested outdoors to determine their impact on photovoltaic power production. Test results showed that PV power diminished to 14% and 22% for HPPSG and HPPSP respectively, compared to a PV panel alone. 2005 Elsevier Ltd. All rights reserved. Keywords: Photodegradation; Solar energy; Water treatment
1. Introduction The most common solar energy applications are thermal, photovoltaic (PV), and photochemical energy conversion. Over the last few years, there has been a growing interest in photovoltaic/thermal collectors
* Corresponding author. Tel.: +41 21 693 4720; fax: +41 21 693 4722. E-mail address: cesar.pulgarin@epfl.ch (C. Pulgarin).
(Sorensen and Munro, 2000), which can provide both electrical and thermal energy from the same system. Nevertheless, there are no reports of photovoltaic/photocatalytic hybrid systems. Fig. 1 (King and Kratochvil, 1997) shows the spectral response of a typical crystalline silicon (c-Si) photovoltaic module compared to the solar spectral irradiance. About 40% of the solar energy incident on a c-Si cell passes through it, because its energy is lower than that of the c-Si band gap. However a major part is absorbed by the sealing and other protective materials used in the
0038-092X/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.solener.2005.02.022
354
V. Sarria et al. / Solar Energy 79 (2005) 353–359
Nomenclature HPPS hybrid photocatalytic–photovoltaic system HPPSG hybrid photocatalytic–photovoltaic system using commercial glass HPPSP hybrid photocatalytic–photovoltaic system using plexiglas
1.8 1.6
Irradiance, (W m-2 nm-1)
1.4 Solar spectra
1.2 1 0.8 0.6 PV cell
0.4 0.2 0 305
410
570
780
980
1395
1800
2537
Wavelength (nm)
Fig. 1. Relative spectral response of a typical c-Si module compared to total solar spectral irradiance (King and Kratochvil, 1997).
module. This may increase the temperature of the module to around 80 C under hot humid conditions. Warming of the PV module can adversely affect its performance due to deterioration of its opto-electronic properties. Moreover, it is known (Custodio et al., 2001) that the short circuit current (Isc) increases with the temperature, whereas the open circuit voltage (Voc) decreases in the same conditions. Generally, the decrease in voltage is much more pronounced than the increment in current (Radziemska, 2003). From a microscopic point of view, the temperature increment in the PV cell causes decreased lifetime and diffusion length of the minority carriers, as well as a decrease in mobility of the majority carriers. Also, when carriers are generated on both sides of the junction, both Fermi levels tend toward the intrinsic level, thus decreasing the potential barrier. The decrease in barrier potential causes an increase in short circuit current and a decrease in open circuit voltage. Efficiency usually falls 30% when the temperature of the cell increases from 20 to 60 C.
PV UV VIS IR CPC
photovoltaic system ultraviolet solar radiation visible solar radiation infrared solar radiation compound parabolic collector
Custodio et al. (2001) had designed an infrared water filter to absorb the part of the solar spectrum, which cannot be exploited by the photovoltaic module and is detrimental for the ideal performance of the module. This filter consisted of a layer of purified water, which is almost transparent to that part of the solar spectrum (400–800 nm) to which c-Si is highly sensitive, and highly absorbent for infrared wavelengths outside of this spectral range. Photocatalytic systems: In recent years, solar UV/visible radiation has been successfully applied for water treatment in the context of so called Advanced Oxidation Processes (AOPs) (Ollis and Ekabi, 1993). These methods use the interaction between ultraviolet radiation and a catalyst or oxidant, to generate OH radicals in the bulk of the solution. OH radicals have a very high oxidation potential and are able to degrade organic compounds present in water, generating non-toxic, biodegradable intermediates that can be removed in a subsequent biological treatment process (Sarria et al., 2002, 2003). Blake (2001) has compiled a survey of literature describing photocatalysis for removal of hazardous compounds from water and air listing 941 scientific papers that describe a variety of process applications. Among the AOPs, the photo-assisted Fenton Fe(III)/ light system, without addition of any electron acceptor other than O2 from air, has received special attention in earlier work, as a potential wastewater pre-treatment (Mailhot et al., 2002). Previous research has shown that under UV irradiation, iron(III) salts can promote the photooxidation of organic compounds (Mailhot et al., 2000; Mazellier and Bolte, 2001; Rodriguez et al., 2002). The agent responsible for these reactions could be the hydroxyl radical formed by photochemical dissociation of Fe(III)-hydroxy complex (Fe(OH)2+). The reaction can be simply expressed as (Faust and Hoigne, 1990): FeðOHÞ2þ þ hm ! Fe2þ þ OH. Since the silicon PV panel converts mainly visible solar radiation into electricity, and since the photocatalytic degradation needs near UV radiation, the approach of simultaneously employing both types of radiation in one engine and with no additional surface can be advantageous.
V. Sarria et al. / Solar Energy 79 (2005) 353–359
Consequently, the aim of this work was to develop a hybrid photocatalytic–photovoltaic system (HPPS). The photocatalytic system uses the UV radiation to promote the degradation of organic pollutants and absorbs the IR radiation, thereby protecting the PV panels and lengthening their lifetime. The photovoltaic system converts sunlight into electricity which can be transformed to alternating current for powering the recirculation pump. It could also be accumulated by a battery for other purposes. This new concept can therefore represent an autonomous and environmentally friendly method for the treatment of biorecalcitrant pollutants. In this paper, we report the construction and the results of basic batch experiments using the HPPS. Photocatalytic degradation of a model biorecalcitrant compound, 5-amino-6-methyl-2-benzimidazole (AMBI) was performed, and the efficiency of the HPPS treatment was compared with a compound parabolic solar collector. In order to validate our results, each experiment was performed three times, and standard deviation was calculated and reported for Table 2 and Figs. 5 and 6.
2. Experimental section 2.1. Hybrid photocatalytic–photovoltaic system (HPPS) A schematic representation of the new HPPS is presented in Fig. 2. Solar radiation comprising the UV, Vis and IR radiations, reaches the photochemical reactor which uses the UV radiation for degradation of the organic pollutants. The IR radiation is filtered by the water in the photocatalytic reactor and only the visible
355
radiation, useful for the production of electrical energy, reaches the PV panel situated below the photocatalytic reactor. Fig. 3 shows a picture of the HPPS installed at the EPFL—Switzerland. It is made up of 4 modules (collector surface, 1.5 m2, photoreactor volume 7 l, and total reactor volume 25 l) mounted on a fixed platform inclined 46 (local latitude). Each module consists of one photocatalytic reactor and one PV panel. The photocatalytic system consists of 4 rectangular cells made of a UV/Vis transparent material through which the water to be treated recirculates. These cells, placed over the PV panels, are connected in series and the water flows directly from one to another and finally to a tank equipped with a centrifugal recirculation pump. The photovoltaic system, built in collaboration with Solstist SA., is composed of the following parts: • 4 PV panels, model BP 270 F (50-W) connected in series; • A sinewave inverter (Studer Joker 400), which converts the low voltage DC to 220-V AC, and stops energy accumulation in the battery; • A battery (12 V 105 Ah) to store the energy; • A Voltmeter to measure the voltage in the battery; • A battery charger, as the solar energy is insufficient for a practical application; • Two electricity meters, to enable the difference between the electricity from the PV panels to be compensated from the local electricity grid; • Two chronometers, for timing the PV/local grid power input;
Fig. 2. Schematic representation of the HPPS.
356
V. Sarria et al. / Solar Energy 79 (2005) 353–359
Fig. 3. Photo of the HPPS installed at the EPFL—Switzerland.
• A three-position commuter (1. automatic, 0. local electricity network, and 2. solar PV panels). 2.2. Compound parabolic collector The low concentrating CPC collector (with a concentration ratio of about 1), described in a previous publication (Sarria et al., 2003) consists of three 4-tube modules (photoreactor material: Pyrex, collector surface: 1.5 m2, photoreactor volume: 12 l, and total reactor volume: 25 l). The three reactor modules are mounted on a fixed platform inclined 46 (local latitude) and connected in series with water flowing directly through them at 30 l min1, leading finally to a recirculation tank connected to a centrifugal pump. Solar ultraviolet radiation (UV), which is a highly significant parameter for the correct treatment of data obtained from solar photocatalytic experiments, was determined during the experiments. A global UV KIPP & ZONEN, model CUV3 radiometer mounted on a 46 fixed-angle platform (the same angle as the CPCs) was used to provide data on global solar energy incident per unit area, WUV m2 (UVG,n). Variation in solarUV power during experiments is made still more noticeable by clouds. With Eq. (1), combination of data from several days experiments and their comparison with other photocatalytic experiments is possible. QUV;n ¼ QUV;n1 þ Dtn UVG;n
Areactor V TOT
ð1Þ
where Dtn = tntn1, tn is the experimental time for each sample, UVG;n is the average UVG during Dtn, Areactor is the irradiated surface of collectors CPC or HPPS (1.5 m2), VTOT is the total volume of the photoreactor (25 l), and QUV,n is the accumulated incident energy on the reactor for each sample during the experiment per unit of volume (kJ l1).
2.3. Reagents 5-amino-6-mehtyl-2-benzimidazolone (AMBI), a previously described (Sarria et al., 2002) biorecalcitrant dye (C8H9N3O), was chosen to test photocatalytic degradation. FeCl3 Æ 6H2O was purchased from Fluka and was used without further purification. The chemicals purchased to assist in HPLC analyses were obtained from Fluka. Milli-Q water was used throughout for preparation of aqueous solutions or as a component of the water-acetonitrile (HPLC grade) mobile phase in HPLC analysis. 2.4. Chemical analyses High performance liquid chromatography (HPLC) was carried out in a Varian 9065 unit using a diode array detector between 200 and 400 nm. A Spherisorb silica column ODS-2 was used and the mobile phase (flow: 1 ml min1) consisted of an acetonitrile/water mixture (50%/50% v/v). This technique allows measuring the AMBI concentration in solution. 3. Results 3.1. Transmittance of different materials suitable for manufacture the HPPS To optimize the performance of the HPPS, the photocatalytic reactor must be made of highly UV-transmissive material. Fig. 4 shows the UV/Vis transmissivity spectra for different materials. Quartz and Pyrex have the highest UV transmissivity; but are not suitable for this purpose, due to their high cost. Commercial glass and Plexiglas, on the other hand, are economically suitable and also resistant to pressure and heating during outdoor operation.
V. Sarria et al. / Solar Energy 79 (2005) 353–359
radiation as shown on the right of Fig. 5. The power is observed to be only slight diminished by the photocatalytic reactor: 14% and 22% by the HPPSG and HPPSP, respectively. Another important parameter of the PV panel is the conversion efficiency. The average efficiency calculated for the PV panel is about 5.8% under normal operative conditions (without photoreactor) and about 4.89% and 4.67% for HPPSG and HPPSP. The above shows that the energy conversion of the PV panel is not significantly diminished by the installation of a photocatalytic reactor over the PV panel.
100 90
Quartz
Transmissivity (%)
80 70
Pyrex
60 50 Commercial glass
40 30 20
Plexiglas
10 0 200
300
400
500
600
357
700
800
3.3. Comparison of a CPC and the new HPPS system for the photocatalytic degradation of AMBI
Wavelength (nm) Fig. 4. Transmissivity of different materials suitable for manufacture of the HPPS.
Two photocatalytic reactors, one using commercial glass (HPPSG) and the other using Plexiglas XT (HPPSP) were fabricated for these experiments. 3.2. Photovoltaic power The sunlight-to-electric conversion parameter to be maximized is the power, which is the product of: current (I) · voltage (V) measured in a photovoltaic module. Fig. 5 shows the photovoltaic power of a PV panel, with and without glass and Plexiglas photocatalytic reactors as a function of the local time between 12:00 and 17:00. Power decreases as time increases, which is to be expected taking into account the diminishing solar
25
Power (W)
20
400
irradiance
HPPSG
300
HPPSP
15
200
10
Irradiance (W m-2)
PV
100
5 0 h0 12
0 h0 13
0 h0 14
0 h0 15
0 h0 16
0 h0 17
Local time
Fig. 5. PV panel power: (PV) panel alone, (HPPSG) with a commercial glass photoreactor, and (HPPSP) with a Plexiglas photoreactor in the new HPPS system. Comparison with the solar irradiance at the EPFL—Switzerland, October 7th 2002.
In order to evaluate the photocatalytic performance of the new HPPS, we simultaneously performed two degradation experiments using a well-known biorecalcitrant pollutant, AMBI (Sarria et al., 2001), one in the new HPPS reactor and another in the CPC, which until now has been demonstrated to be the best solar photocatalytic reactor for degradation applications (Malato et al., 2002), and compared the results. A diluted solution of AMBI was degraded by photoassisted Fenton using the CPC, the HPPSG and the HPPSP. The main characteristics of the reactors are presented in Table 1. Both the glass and Plexiglas HPPS reactors have the same dimensions. All three systems have the same illuminated surface. The main difference between these systems is the recirculation rate, which is higher in the CPC. The transversal section for photoreactors in the CPC and HPP systems are different. For that reason, the water flow for both systems was experimentally estimated using the ‘‘stop watch and bucket method’’. Basically, it was recorded the time to deliver a known volume, then volume divided by time is the flow rate. However, it was decided to go further in the exploratory evaluation of system performance, comparing the same illuminated surface. Fig. 6 shows ln(C/C0) evolution of the AMBI solution as a function of accumulated energy, calculated using Eq. (1), during the photocatalytic degradation experiments with the 3 solar reactors. C is the concentration during the experiment and C0 is the initial
Table 1 Characteristics of the compound parabolic collector (CPC) and the hybrid photovoltaic photocatalytic system (HPPS) Reactor Illuminated Illuminated Total Recirculation area (m2) volume (l) volume (l) (l min1) CPC HPPS
1.5 1.5
12 7
25 25
30 0.8
358
V. Sarria et al. / Solar Energy 79 (2005) 353–359 0.4 0.35 CPC
-ln (C/Co)
0.3 0.25
HPPSG
0.2 0.15
HPPSP
0.1
y = 0.001x
0.05 0 0
25
50
75
100
125
Quv(kJ l-1) Fig. 6. Comparison of the CPC and the HPPS systems.
Table 2 Comparison of the photodegradation constant (kapp) of AMBI solutions, using the HPPS and CPC reactors System
HPPSP
HPPSG
CPC
kapp (l kJ1)
0.0012 (±0.0004)
0.0021 (±0.0005)
0.0031 (±0.0003)
concentration of AMBI at the beginning of the experiment. According to the results presented in Fig. 6, it is clear that the CPC is the more performant from the kinetic point of view, however the HPP systems are energy autonomous. The constant of photodegradation (kapp) is obtained by calculating the slope of the curves in Fig. 6. Table 2 illustrates the kapp obtained for each reactor. Under the experimental conditions, the CPC is 1.5 and 3.1 times more efficient than the HPPSG and HPPSP, respectively for the treatment of a diluted solution of AMBI. However, more research will be necessary to find out whether the HPPS efficiency can be improved. The parameters involved should be, for instance, another kind of UV-transmissive material, agitation in the HPPS water circuit, water recirculation rate, and the use of an immobilized catalyst.
4. Conclusions A new hybrid photocatalytic photovoltaic system (HPPS) was developed, installed and operated at the EPFL in Switzerland. This HPPS uses solar UV radiation to evoke the photocatalytic degradation of organic pollutants, and visible radiation to generate electricity needed by the recirculation pump. HPPS represent an autonomous and environmental friendly approach for the treatment of biorecalcitrant pollutants. Commercial glass and Plexiglas, which have a good transmittance-cost ratio, were used to manufacture two
HPPS prototypes, HPPSP (Plexiglas) and HPPSG (commercial glass). The photovoltaic power of the PV panel was compared with and without the glass and Plexiglas photocatalytic reactors. The power output of the cell is almost directly proportional to the intensity of the sunlight. It was also observed that the power is only slightly diminished by addition of the photocatalytic reactor: 14% and 22% with the HPPSG and HPPSP, respectively. Degradation of the model compound, 5-amino-6methyl-2-benzimidazol (AMBI) is possible using photo-assisted Fenton. Under the experimented conditions, the CPC was shown to be 1.5 and 3.1 times more efficient than the HPPSG and HPPSP, respectively, for the degradation of AMBI. Acknowledgements This work was funded by OFES Contract no: 01.0433 under the European Community CADOX Research Project Contract no: EUK1-CTIC1-CT-200200122. We wish to thank Pascal Afolter of ‘‘Solstis SA’’. V. Sarria thanks ‘‘Comite´ de Investigaciones y Posgrados de la Facultad de Ciencias, Universidad de los Andes’’ for the financial support. We also wish to express our gratitude to the ‘‘Project environment EPFLUNIVALLE’’ and the skillful technical assistance of Jean-Pierre Kradolfer. The authors also wish to thank Mrs. Deborah Fuldauer for correcting the English. References Blake, D., 2001. Bibliography of work on the heterogeneous photocatalytic removal of hazardous compounds from water and air. Update number 4 to October 2001, NREL/ TP-510-31319. National Technical Information Service (NTIS), U.S. Department of Commerce, Springfield, VA 22161. Custodio, E., Acosta, L., Sebastian, P.J., Campos, J., 2001. A better solar module performance obtained by employing an infrared water filter. Solar Energy Materials and Solar Cells 70, 395–399. Faust, B.C., Hoigne, J., 1990. Photolysis of Fe(III)-hydroxy complexes as sources of OH radicals in clouds, fog and rain. Atmos. Environ. 24A, 79–89. King, D.L., Kratochvil, J.A., 1997. Measuring solar spectral and angle-of-incidence effects on photovoltaic modules and solar irradiance sensors. In: Proceedings of the 26th IEEE Photovoltaic Specialists Conference. Sandia National Laboratories, Anaheim, California. Mailhot, G., Asif, A., Bolte, M., 2000. Degradation of sodium 4-dodecylbenzenesulphonate photoinduced by Fe(III) in aqueous solution. Chemosphere 41, 363–370. Mailhot, G., Sarakha, M., Lavedrine, B., Caceres, J., Malato, S., 2002. Fe(III)-solar light induced degradation of diethyl phthalate (DEP) in aqueous solutions. Chemosphere 49, 525–532.
V. Sarria et al. / Solar Energy 79 (2005) 353–359 Malato, S., Blanco, J., Vidal, A., Fernandez, P., Caceres, J., Trincado, P., Oliveira, J.C., Vincent, M., 2002. New large solar photocatalytic plant: set-up and preliminary results. Chemosphere 47, 235–240. Mazellier, P., Bolte, M., 2001. 3-Chlorophenol elimination upon excitation of dilute iron(III) solution: evidence for the only involvement of Fe(OH)2+. Chemosphere 42, 361–366. Ollis, D.F., Ekabi, H.A. (Eds.), 1993. Photocatalytic Purification and Treatment of Water and Air. Elsevier, Amsterdam. Radziemska, E., 2003. The effect of temperature on the power drop in crystalline silicon solar cells. Renewable Energy 28, 1–12. Rodriguez, M., Abderrazik, N., Contreras, S., Chamarro, E., Gimenez, J., Esplugas, S., 2002. Iron(III) photoxidation of organic compounds in aqueous solutions. Appl. Catal. B. 37, 131–137.
359
Sarria, V., Parra, S., Invernizzi, M., Pe´ringer, P., Pulgarin, C., 2001. Photochemical–Biological Treatment of a Real Industrial Biorecalcitrant Wastewater Containing 5-amino-6methyl-2-benzimidazolone. Water Sci. Technol. 44, 93–101. Sarria, V., Parra, S., Adler, N., Pe´ringer, P., Pulgarin, C., 2002. Recent developments in the coupling of photoassisted and aerobic biological processes for the treatment of biorecalcitrant compounds. Catal. Today. 76, 301–315. Sarria, V., Deront, M., Peringer, P., Pulgarin, C., 2003. Degradation of a biorecalcitrant dye precursor present in industrial wastewaters by a new integrated iron(III) photoassisted-biological treatment. Appl. Catal. B. 40, 231– 246. Sorensen, H., Munro, D., 2000. Hybrid PV/T Collectors. In: Proceedings of the 2nd World Solar Electric Buildings Conference, Sydney, Australia.